Selenophosphate (SeP), the energy-rich Se compound required for synthesis of many selenoenzymes is formed by selenophosphate synthetase (SPS) from ATP and an inorganic form of Se. The participation of specialized selenium delivery proteins to furnish Se specifically to SPS serves to maintain intracellular levels of selenium below toxic levels. Certain classes of delivery proteins that can utilize selenocysteine as substrate, selenocysteine lyases, include three NifS related proteins from E. coli and one from Methanococcus vannielii, an anaerobic organism that is particularly rich in selenoenzymes and factors required for selenoprotein biosynthesis. Bovine rhodanese, a known sulfur transferase, can be converted to a stable perselenide adduct and this serves as a Se delivery protein model. Additional proteins that may have roles in selenium transport are under study by Dr. Gerard Lacourciere. Human thioredoxin reductase, a homodimeric enzyme, contains an essential selenocysteine residue located in a tripeptide sequence, -Cys, Secys, Gly, at the C-terminus of each subunit. Expression of this selenoprotein in E. coli is very inefficient and termination at the UGA codon is frequent. To facilitate isolation of full-length active enzyme species, constructs encoding C-terminal extensions that would bind to specific affinity matrices were tested by Dr. Shoshana Bar-Noy. A polyhistidine tag separated by a protease cleavage site allowed selective enrichment of full-length UGA-readthrough products on a nickel affinity column. Subsequent protease cleavage removed the C-terminal tag. Certain bacterial enzymes, such as purine and xanthine hydroxylases that are members of the selenium-dependent molybdenum hydroxylase family, lack selenocysteine and instead contain a labile cofactor form of Se. The immediate source of selenium and its mechanism of insertion into these enzymes is a major research effort of Dr. William Self. The selenotrisulfide adduct of lipoic acid, a potential selenium source for these hydroxylases, is a more stable form of selenium that also is viewed as a potential Se-delivery antioxidant for treatment of radiation-induced skin damage in experimental animals. Collaborative studies with Duke University clinicians are in progress. Follow up of earlier observations of HIV-induced destruction of Se-enzymes in T-cells with accumulation of anionic polyselenide compounds and similar lower level effects mimicked with Tat protein alone is the subject of research by Dr. Lara Campbell. It has been proposed that Tat protein may interfere with selenoprotein biosynthesis by competitive binding of the SECIS stem loop RNA structure, which resembles the HIV mRNA stem loop structure, TAR, the normal target of Tat protein.